Performance Analysis of BitTorrent Protocol
|
|
- Walter Reed
- 6 years ago
- Views:
Transcription
1 Performance Analysis of BitTorrent Protocol Parul Sharma Department of Computer Science Engineering Anuja Bhakuni Department of Computer Science Engineering Rishabh Kaushal Department of Information Technology Abstract BitTorrent protocol has been quite successful in capturing majority of Internet traffic since its inception due to its various mechanisms to improve the download performance of the network. The mechanisms like tit-fortat, optimistic unchoking, anti-snubbing and various piece selection strategy, etc. have been working very well. This work is an attempt to study the performance of BitTorrent protocol and its networking infrastructure through two strategies, namely, (A) Simulations based using BitTorrent patch in ns-2 simulator and (B) Real networks using client implementation based on libtorrent library. Empirical results thus obtained are used to infer various performance parameters of BitTorrent protocols. Keywords BitTorrent protocol, Peer to Peer (P2P), Internet, Networks. I. INTRODUCTION Ever since the development and subsequent release of BitTorrent protocol by its inventor Bram Cohen in July, 21 [1], it has remained a very popular file sharing protocol. In November 24, BitTorrent accounted for a huge 35 percent of all the traffic on the Internet and in 26 the BitTorrent protocol has risen to over 6 percent of all Internet traffic according to British Web analysis firm CacheLogic and 78 % of all the global p2p traffic [2]. As per a recent article in Jan., 212, it is claimed that BitTorrent is accounting for an unbelievable 15 million monthly users [3]. Furthermore, it is estimated that BitTorrent, at any given point of time, has on an average more active users than YouTube and Facebook combined [4] [5]. One key aspect of performance in BitTorrent is the download rate that is achieved by participating peers. In this paper we try to find the relation of download speed with various parameters like upload bandwidth, swarm size, starttime, first-chunk-time etc. It is done by performing p2p simulations through NS2 simulator and also through libtorrentrasterbar in the real environment. The remainder of paper is organized as follows. Working of the BitTorrent protocol is explained in Section II. Various mechanisms like peer and piece selection strategies and the historic work done are presented in section III and section IV respectively. In Section V, we present our empirical analysis of BitTorrent network. Finally conclusion is drawn in section VI. II. WORKING OF BITTORRENT PROTOCOL BitTorrent protocol was primarily designed aiming to be a substitute of old centralized HTTP downloads, and not a full p2p protocol. But as it improved, it became peer-to-peer in nature where users connected to each other directly to upload and download portions of a large file (called as a piece) from other peers who have also downloaded either the file or parts of it. These pieces are, henceforth, reassembled into the full file. This mechanism provided faster download whereas in normal p2p systems, the peer downloads from a single peer alone and the download speed is limited. The conventional centralized server based form of downloads were marred with problems of server bottlenecks and also contributed to network congestion. While BitTorrent was designed to provide a faster download and congestion free mechanism. The property of simultaneous uploading while downloading ensures that there is no single source to download from (thereby removing the bottleneck problem) and also ensures that network traffic due to file download gets distributed across the network (thereby preventing the chances of network congestion)[6]. To start a BitTorrent deployment, a.torrent metadata file is put on an ordinary web server/portal. The.torrent consists of information about the file, its name, length, URL of a tracker and hashing information. Tracker helps downloader(s) find each other. A very basic protocol called HTTP tracker protocol is used in which a downloader sends a message including information about downloading file, the port used etc. and the tracker replies with a list of peers (including contact information) downloading the same file. Downloader(s) then use this information to connect to each other. To make a file available, a seed, i.e., a downloader which already has complete file starts first. The bandwidth /13/$ IEEE
2 requirements of the tracker (which is a server that maintains a list of clients forming the BitTorrent swarm associated with given content and is also aware of the download progress of each peer within the swarm) and web server are very low, while the seed must send out at least one complete copy of the original file. For downloading the files that are hosted using BitTorrent users must have a BitTorrent client/software (a computer program designed for p2p file sharing using the BitTorrent protocol). The leechers, not having all pieces of a large file, have to upload the pieces it has, and download the ones requested from either leechers or seeders to reassemble these into a complete file. All these users (seeders & leechers) together combine to form a swarm. The download bandwidth load is divided among many sources since the users download from several other users and not from a single central server. This benefits not only the downloader by increasing its download rates but also effectively reduces the cost of upload bandwidth for servers hosting large files, since BitTorrent protocol effectively distributes the upstream bandwidth of a single server across all downloaders, hereby gaining advantage from peers for the whole BitTorrent system. III. MECHANISMS USED IN BITTORRENT PROTOCOL The mechanisms employed by BitTorrent mainly consist of peer and piece selection strategies [7][8]. A superlative peer selection strategy must maximize the service capacity of the system and a good piece selection strategy must ensure that each peer can find pieces that it requires from its neighbors. They play a major role in determining the downloading experience of any peer. A. Peer Selection Strategy: The peer selection strategy of BitTorrent uses four principal mechanisms: tit-for-tat, optimistic unchoking, upload only and anti-snubbing. Collectively, they form the base for the choking algorithm utilized by a peer. The main aim of these mechanisms is to improve the downloading performance of those peers that help induce the file exchange by uploading pieces to other peers, and punish freeriders. i. Tit-for-tat: The tit-for-tat strategy is taken up by a peer to preferentially upload to its neighbor peers who provide it the optimum downloading rates. At any given time, a peer exchanges file pieces with a fixed number of neighboring peers. A leecher preferentially uploads to the best neighbors who provide it the best downloading rate and chokes others. Every few seconds, the leecher reevaluates the downloading rate from all peers who are sending data to it. If another peer offers better downloading rate, the leecher will choke the peer with the smallest downloading rate among the current peers it is serving, and unchoke the better peer. This mechanism is very important to encourage downloaders and punish free-riders, thus preventing leechers from downloading without uploading anything. ii. Optimistic Unchoking: BitTorrent incorporates an optimistic unchoking policy, besides the TFT mechanism. Every few seconds, a peer unchokes a neighboring peer randomly regardless of its uploading rate. The benefit of Optimistic Unchoking is that it enables a peer to discover better neighboring peers to exchange file iii. iv. pieces with since other neighbors may have higher capacity for uploading than the currently unchoked peers. If a peer uses only the TFT mechanism, there will be no opportunity for discovering other peers that can provide higher uploading rate. This strategy is extremely useful for newly joined peers to get started. Upload Only: A peer becomes a seed, once it finishes downloading the entire file. Seeds cannot select peers based on downloading rates, as seeds have nothing to download. The seeds prefer to upload to peers with better uploading rates. Anti-snubbing: A peer may be choked by the peers it was earlier downloading from, thereby getting poor downloading rates. To address this problem, when a peer notices that some time has elapsed without getting a single piece from a neighboring peer, the leecher assumes it is snubbed by that peer and does not upload to it any further through regular unchoke. B. Piece Selection Strategy: The second main strategy that plays a significant role in the success of BitTorrent is the piece selection strategy. When a peer wants to download pieces from its neighbours, it adopts several piece selection strategies, those mechanisms are as follows: i. Strict Priority: In BitTorrent, peers concentrate on downloading a whole piece of a file, before requesting for another piece.thus if a sub-piece is requested, then subsequent subpieces of the same piece will be requested preferentially so as to complete the download of the whole piece as soon as possible, as only complete pieces can be exchanged with others. ii. Endgame Mode: This mode is taken up by a peer in the end of downloading the file. If a piece is requested from a peer that has a slow transfer rate, the downloading time gets prolonged. To address this problem, a peer requests all of its neighbours for blocks it has not received. After a block is obtained, the peer aborts the request for that block from its neighbours, so as to lessen the bandwidth wastage due to redundant downloads. iii. Rarest First: Peers generally prefer to download pieces that are the rarest among their neighbours. This strategy is known as the rarest first algorithm which works as follows. Every peer maintains a list of the pieces that each of its neighbours contains. This list is updated every time a copy of a piece becomes available from its neighbours. It then creates a rarest-pieces set which is the list of those pieces that have the minimum number of copies among its neighbours. It then chooses the piece to download that is rarest among its neighbours. This strategy increases the chance of a peer exchanging with its neighbours as it has pieces that others require. This also benefits the seeds by reducing their server burden, since they need to upload less copies of the same piece. Furthermore, the exit of a seed will not leave the remaining
3 iv. downloaders without the opportunity to exchange file pieces, because this strategy assures that all the pieces are quickly distributed to maximum of the leechers. Random First Piece: An exception exists to the rarest first policy when a peer first joins a torrent. Since the peer does not have any pieces, so it would like to download a whole piece quickly so that it can get ready to reciprocate with the TFT algorithm. In this case, the new peer will not take the rarest piece. Instead, it will download the first piece randomly in order to make sure it can get a whole piece as soon as possible. IV. RELATED WORK A lot of studies have been performed on the measurements and modeling BT performance affected by various parameters. Izal et al. [9] presented the first comprehensive evaluation of BitTorrent using measurement and used the tracker log of Linux RedHat 9 distribution for their evaluation. Here, they indicated that the uploading and downloading rates are correlated, which verified the effectiveness of TFT policy. Differing from Izal [1] testing only on a single torrent, Legout et al. [11] conducted a detailed measurement study aimed at understanding the behavior of the rarest-first algorithm and the choking algorithm on several real torrents. Chih-Lin HU et al. [12] investigated on a peer-side basis in terms of downloading side bandwidth allocation and usage with various parameters like bandwidth allocation and file size for real traffic traces. Bharambe et al. [13] focused on two main metrics: utilization of upload capacity and (un)fairness in terms of data size served by node and hence evaluated the impact of BitTorrent s mechanisms on swarm performance under a variety of flashcrowd workloads using a simulator, which models both peer s activity and BitTorrent s mechanisms like TFT and rarestfirst. Felber et al. [14] presented the effect of different peer and piece selection strategies on the performance of BitTorrent using simulations Urvoy-Keller et al. [15] shows the significant impact of two overlay topology parameters, peer set size (PS) and percentage of outgoing connections(oc) on BitTorrent s performance using a simulation approach. Differing from the studies above where the focus is on upload efficiency and overall system performance, we analyze the download performance of a single user (rather than a whole swarm) against parameters like upload rate, number of peers and first chunk time using ns2 simulator and real-time libtorrent client to verify various piece and peer selection strategies affecting performance of the downloader. V. EMPIRICAL ANALYSIS OF BITTORRENT NETWORK In this section, we present an extensive empirical analysis of the BitTorrent protocol and its supporting networking infrastructure. Two strategies have been adopted for the empirical analysis, namely, (A) Simulation based analysis, which was done using ns-2 simulator [15][16] and (B) Realnetwork analysis which was done using client example provided by the open-source libtorrent library compiled with qtorrent option as GUI. A. Simulations using ns2 We used ns2 simulator [15] to test the BitTorrent protocol and the effects that the chosen parameters have on the download time performance of swarm. To facilitate our analysis, we have used bt_flashcrowd_star.tcl (provided by [16]) file to create a BT network which essentially checks the efficiency of data transfers between peers, ignoring checking the download data integrity by hash values. We chose the ns2 based BitTorrent implementation since it supports various input-output parameters and is also easy to understand and simple to use. As per this implementation, unchoking algorithm is implemented while anti-snubbing and endgame features of BT protocol were ignored. Super-seeding which improves flash crowd scenario has been implemented. The peer and piece selection algorithm can be replaced by alternatives being modular. The ns-2 simulator was patched with BitTorrent (BT) patch which creates 4 additional classes to support basic BitTorrent protocol operation, namely, BitTorrent Application, BitTorrent Tracker, BitTorrent Connection and BitTorrent message. BT patch uses full duplex TCP (bidirectional) as the carrier transport layer protocol for simulating BitTorrent traffic using ns-2 simulator. Following three input parameters which were tested to verify the outcome: i. Number of peers. ii. Seed with random number. iii. Download Rate of the network (in ). The output log file generated by ns-2 simulator contained the following output parameters which were verified against input parameters: i. Peer ID, unique ID of the peer. ii. Start Time, time when the peer joins the BitTorrent network. iii. First Chunk Time, time when the peer receives its first piece of file. iv. Download Finish Time, time when the peer completes its download. v. End Time, time when the entire swarm finishes downloads. vi. Total Download Time, is calculated as the difference between Download Finish Time and Start Time. Keeping number of peers as constant (configured to a value of 1), analysis was done to ascertain when the peer gets its first chunk with respect to the time when it started its download. Download rate was varied during the run with different values - 5, 1, and 2. As evident in Figure 1, it is observed that the first chunk time of the fastest network rate (i.e. = 2 ) was the least, that is the peers got their first piece in the fastest time. But varying this rate to 1 didn t have much effect on the first chunk time. But when the rate dropped down to 5, the first chunk time increased considerably. Conclusion: Keeping download rate fixed, the time when the peer receives its first piece of file (chunk) is proportional to when it joins the network (start time) i.e. the first chunk time increases with rise in start time during the initial run of network when the load is less, however, afterwards, this trend no longer exist when the network load increases.
4 Download Finish Time {s} Download Finish Time for swarm (s) First Chunk Time (s) Download Finish Time (S) Start Time (s) Figure 1: Start Time vs. First Chunk Time (# peers = 1) In the next analysis, download finish time was compared with the start time for each peer (again keeping number of peers fixed to 1) with varying download rate of 5, 1 and 2, as show in Figure 2. Conclusion: It is observed that that the download finish time for each peer remains constant, in other words, it can be concluded that it is independent of download start time (keeping download rate constant). It is due to the fact that peers joining late in the network have more file pieces distributed across multiple peers available for parallel downloads. However, by increasing the download rate from 5 to 2, the Download Finish Time of a single peer decreases. Conclusion: Download Finish Time improves (reduces) by increasing download rate. First Chunk Time (s) Figure 3: First Chunk Time vs. Total Download Time The Download Finish Time for the swarm is verified against number of peers present in the network with download rates ranging from 5, 1 and 2 in Figure 4. For a constant download rate, say 2, the Download Finish Time increases linearly with number of peers. This trend is attributed due to the fact that network load increases as more peers join the network, thereby increasing the download finish time. In addition, when rate is doubled from 5 to 1, the Download Finish Time also doubles, when viewed from the point of view that number of peers remains same, say Start Time (s) Number of peers Figure 2: Start Time vs. Download Finish Time (#peers = 1) Next, Total Download Time was studied with respect to the Start Time keeping different values of download rates. In this case, observations are in consonance with the earlier ones. Conclusion: The Total Download Time of each peer remains same as long as download speed is not varied. Furthermore, even Total Download Time is invariant to the Start Time and increases with an increase in speed of the network. The Download Finish Time (Figure 3) and Total Download Time were verified against the First Chunk Time, with varying download rate of 5, 1, and 2. Similar to the Start Time analysis, it was observed that First Chunk Time doesn t have significant effect on the Download Finish Time and the Total Download Time, which more or less remained constant within the same download rate level (say 484 when speed = 5 ) and increases approx. 1 % every time speed is doubled. Figure 4: Number of peers vs. Download Finish Time B. Real time analysis using LibTorrent library Open source libtorrent library was compiled with qtorrent option as GUI enabled and its client example binary was used to download actual files over the Internet in order to observe the behavior of BitTorrent protocol in real world. To measure the effect of factors affecting the performance efficiency, libtorrent library was compiled, in particular the an example program called client_test, which implements a complete BitTorrent client was built. Subsequently, it was used to download numerous files from Internet by passing the requisite torrent file via command line. This section presents a representative view of the observations made during the runs of such downloads. Care was taken to download with relevant parameters which affected the download rate of the file. Particularly, the download performance was checked against two parameters: i) Upload rate. The rate at which sub-pieces of a downloaded file is provided to other peers. The downloading rate was
5 Download Rate () Download Rate () checked against varying upload rates. Figure 5 shows the graph depicting the effect of upload rates on download rate of torrent file. Here, the download rate is taken as average. The average number of peers was taken as 5 within a swarm. Conclusion: Interestingly, it was observed that increasing the upload speed increases the download speed as well, a verification of tit for tat strategy, however, after a certain upload rate threshold, it is observed that the download rate decreases Upload Rate() Figure 5: Upload Rate vs. Download Rate ii) Number of peers connected, also known as swarm size. The peers in connection selected by Tracker, form a group that helps in downloading file to each other. The downloading speed was checked under different swarm size. Here also, the download rate is taken as average. Figure 6 shows the graph depicting this table. Conclusion: Although, number of peers does help in download rate considerably due to sharing of file pieces, however, beyond a point, it results in decrease in download rate (speed) Number of peers Figure 6: Number of Peers vs. Download Rate VI. CONCLUSION We studied in this paper the working of BitTorrent Protocol and various mechanisms used to achieve optimal performance by the protocol. In this paper we investigated the impact of number of peers on download rate through NS-2 simulator and libtorrent client. It was found that the Total Download Time decreases when number of peers increase due to an increased availability of file pieces across multiple peers. However, Download Time increases beyond certain limit due to increase of traffic in the network. Another inference drawn from NS-2 based analysis is that howsoever late a peer joins the network, it finishes in almost the same time as compared to peers joining early. Based on libtorrent based analysis, it was found that download rate is maximum when the upload rate is configured neither too low nor too high. This practical trend is in compliance to the Tit-for-Tat strategy of BitTorrent protocol. In this regard it is further concluded that BitTorrent incentive mechanism is playing a key role in peer-to-peer downloads. REFERENCES [1] Cohen, Bram (21-7-2). "BitTorrent a new P2P app". Yahoo egroups, 21. [2] Hendrik Schulze, Klaus Mochalski, Internet Study 28/29, Ipoque, 29. [3] "BitTorrent and µtorrent Software Surpass 15 Million User Milestone", Bittorrent.com, Jan., 212. [4] BitTorrent Has More Users Than Netflix and Hulu Combined-- and Doubled, Jan., 211. [5] Online Video Rankings, Aug., 21. [6] Michal Kryczka, Rubén Cuevas, Carmen Guerrero, Arturo Azcorra, Angel Cuevas, Measuring the BitTorrent Ecosystem: Techniques, Tips, and Tricks, Communications Magazine IEEE [7] B. Cohen, Incentives build robustness in BitTorrent, in First Workshop on Economics of Peer-to-peer Systems, Berkeley, USA, June 23 [8] Raymond Lei Xia and Jogesh K. Muppala, Senior Member, IEEE, A Survey of BitTorrent Performance [9] M. Izal, G. Urvoy-keller, E. W. Biersack, P. A. Felber, A. A. Hamra, and L. Garces-Erice, Dissecting BitTorrent: Five months in a torrents lifetime, in Passive and Active Measurements (PAM), LNCS, vol. 314, April 24, pp [1] A. Legout, G. Urvoy-Keller, and P. Michiardi, Rarest first and choke algorithms are enough, in IMC 6: Proc. 6th ACM SIGCOMM conference on Internet measurement. New York, NY, USA: ACM, 26, pp [11] Chih-Lin HU* and Zong-Xian LU, A Trace Study of BitTorrent P2P File distribution with Downloading-Side Performance Measurement and Analysis, Advanced Communication Technology (ICACT), th International Conference on, Feb. 212, pp [12] A. R. Bharambe, C. Herley, and V. N. Padmanabhan, Analyzing and improving a BitTorrent networks performance mechanisms, INFOCOM th IEEE International Conference on Computer Communications Proceedings, pp. 1 12, April 26. [13] P. A. Felber and E. W. Biersack, Self-scaling networks for content distribution, in Int. Workshop on Self-* Properties in Complex Information Systems, Bertinoro, Italy, May-June 24 [14] G. Urvoy-Keller and P. Michiardi, Impact of inner parameters and overlay structure on the performance of BitTorrent, INFOCOM th IEEE International Conference on Computer Communications. Proceedings, pp. 1 6, April 26 [15] NS2 Simulator [16] Kolja Eger, Tobias Hoßfeld, Andreas Binzenh ofer, Gerald Kunzmann, Efficient Simulation of Large-Scale P2P Networks: Packet-level vs. Flow-level Simulations, Proceedings of the ACM SIGCOMM second workshop on Use of P2P, GRID and agents for the development of content networks, pp. 9-16
Content Distribution and BitTorrent [Based on slides by Cosmin Arad]
ID2210 - Distributed Computing, Peer-to-Peer and GRIDS Content Distribution and BitTorrent [Based on slides by Cosmin Arad] Today The problem of content distribution A popular solution: BitTorrent Underlying
More informationBitTorrent. Masood Khosroshahy. July Tech. Report. Copyright 2009 Masood Khosroshahy, All rights reserved.
BitTorrent Masood Khosroshahy July 2009 Tech. Report Copyright 2009 Masood Khosroshahy, All rights reserved. www.masoodkh.com Contents Contents 1 Basic Concepts 1 2 Mechanics 3 2.1 Protocols: Tracker and
More informationScalability of the BitTorrent P2P Application
Scalability of the BitTorrent P2P Application Kolja Eger, Ulrich Killat Hamburg University of Technology 5.Würzburger Workshop 8.-9. July 2005 Overview File dissemination in peer-to-peer (p2p) networks
More informationTHE PEER-TO-PEER (P2P) paradigm has proved to be
140 IEEE COMMUNICATIONS SURVEYS & TUTORIALS, VOL. 12, NO. 2, SECOND QUARTER 2010 A Survey of BitTorrent Performance Raymond Lei Xia and Jogesh K. Muppala, Senior Member, IEEE Abstract Since its inception,
More informationIntroduction to P P Networks
Introduction to P P Networks B Sc Florian Adamsky florianadamsky@iemthmde http://florianadamskyit/ cbd Internet Protocols and Applications SS B Sc Florian Adamsky IPA / Outline Introduction What is P P?
More informationExtreme Computing. BitTorrent and incentive-based overlay networks.
Extreme Computing BitTorrent and incentive-based overlay networks BitTorrent Today we will focus on BitTorrent The technology really has three aspects A standard that BitTorrent client systems follow Some
More informationNew Technique to Improve BitTorrent Performance Based on Application Layer Traffic Optimization
International Journal of Computing and Network Technology ISSN 2210-1519 Int. J. Com. Net. Teach. 3, No. 1 (Jan. 2015) New Technique to Improve BitTorrent Performance Based on Application Layer Traffic
More informationCompSci 356: Computer Network Architectures Lecture 21: Overlay Networks Chap 9.4. Xiaowei Yang
CompSci 356: Computer Network Architectures Lecture 21: Overlay Networks Chap 9.4 Xiaowei Yang xwy@cs.duke.edu Overview Problem Evolving solutions IP multicast Proxy caching Content distribution networks
More informationP2P Applications. Reti di Elaboratori Corso di Laurea in Informatica Università degli Studi di Roma La Sapienza Canale A-L Prof.ssa Chiara Petrioli
P2P Applications Reti di Elaboratori Corso di Laurea in Informatica Università degli Studi di Roma La Sapienza Canale A-L Prof.ssa Chiara Petrioli Server-based Network Peer-to-peer networks A type of network
More informationEfficiency of Data Distribution in BitTorrent-Like Systems
Efficiency of Data Distribution in BitTorrent-Like Systems Ho-Leung Chan 1,, Tak-Wah Lam 2, and Prudence W.H. Wong 3 1 Department of Computer Science, University of Pittsburgh hlchan@cs.pitt.edu 2 Department
More informationP2P Content Distribution
, University of Hannover Wolf-Tilo Balke and Wolf Siberski 28.11.2007 *With slides from K. Wehrle (RWTH), A. Bharambe (CMU), P. Rodriguez, P. Chou (MSRC), Chiu (CUHK) Peer-to-Peer Systems and Applications,
More informationMETHODOLOGIES TO IMPROVE THE PERFORMANCE OF A BITTORRENT LIKE PEER TO PEER NETWORK
METHODOLOGIES TO IMPROVE THE PERFORMANCE OF A BITTORRENT LIKE PEER TO PEER NETWORK 1 M.PADMAVATHI, 2 R.M.SURESH 1 Associate Professor, Department of Information Technology, RMD Engineering College, PO
More informationLecture 17: Peer-to-Peer System and BitTorrent
CSCI-351 Data communication and Networks Lecture 17: Peer-to-Peer System and BitTorrent (I swear I only use it for Linux ISOs) The slide is built with the help of Prof. Alan Mislove, Christo Wilson, and
More informationA Trace Study of BitTorrent P2P File Distribution with Downloading-Side Performance Measurement and Analysis
A Trace Study of BitTorrent P2P File Distribution with Downloading-Side Performance Measurement and Analysis Chih-Lin HU* and Zong-Xian LU Department of Communication Engineering, National Central University,
More informationImproving BitTorrent: A Simple Approach
Improving BitTorrent: A Simple Approach Alix L.H. Chow Univ. of Southern California Leana Golubchik Univ. of Southern California Vishal Misra Columbia University Abstract Measurement studies have shown
More informationImpact of Inner Parameters and Overlay Structure on the Performance of BitTorrent
Impact of Inner Parameters and Overlay Structure on the Performance of BitTorrent Guillaume Urvoy-Keller Institut Eurecom, France Email: urvoy@eurecom.fr Pietro Michiardi Institut Eurecom, France Email:
More informationBiToS: Enhancing BitTorrent for Supporting Streaming Applications
BiToS: Enhancing BitTorrent for Supporting Streaming Applications Aggelos Vlavianos, Marios Iliofotou and Michalis Faloutsos Department of Computer Science and Engineering University of California Riverside
More informationBitTorrent Optimization Techniques. (from various online sources)
BitTorrent Optimization Techniques (from various online sources) Announcement No recitation next week! Final review session Next Sunday (5/2) 5-7pm, GHC 4215 Let us know what you want at http://www.doodle.com/6qvsnubhmam2zkxp
More informationBandwidth Trading in Unstructured P2P Content Distribution Networks
Bandwidth Trading in Unstructured P2P Content Distribution Networks Kolja Eger and Ulrich Killat Department of Communication Networks Hamburg University of Technology (TUHH) {eger, killat}@tu-harburg.de
More informationOn Feasibility of P2P Traffic Control through Network Performance Manipulation
THE INSTITUTE OF ELECTRONICS, INFORMATION AND COMMUNICATION ENGINEERS TECHNICAL REPORT OF IEICE On Feasibility of P2P Traffic Control through Network Performance Manipulation HyunYong Lee Masahiro Yoshida
More informationinria , version 1-6 Sep 2006
Rarest First and Choke Algorithms Are Enough Arnaud Legout I.N.R.I.A. Sophia Antipolis France arnaud.legout@sophia.inria.fr G. Urvoy-Keller and P. Michiardi Institut Eurecom Sophia Antipolis France {Guillaume.Urvoy,Pietro.Michiardi}@eurecom.fr
More informationImproving Performance of BitTorrent Network through Incentive Mechanism
San Jose State University SJSU ScholarWorks Master's Projects Master's Theses and Graduate Research Summer 2011 Improving Performance of BitTorrent Network through Incentive Mechanism Mingzhe Li San Jose
More informationDo incentives build robustness in BitTorrent?
Do incentives build robustness in BitTorrent? ronghui.gu@yale.edu Agenda 2 Introduction BitTorrent Overview Modeling Altruism in BitTorrent Building BitTyrant Evaluation Conclusion Introduction 3 MAIN
More informationContent Overlays (continued) Nick Feamster CS 7260 March 26, 2007
Content Overlays (continued) Nick Feamster CS 7260 March 26, 2007 Administrivia Quiz date Remaining lectures Interim report PS 3 Out Friday, 1-2 problems 2 Structured vs. Unstructured Overlays Structured
More informationP2P content distribution Jukka K. Nurminen
P2P content distribution Jukka K. Nurminen 1 V1-Filename.ppt / yyyy-mm-dd / Initials BitTorrent content downloading Efficient content distribution Bram Cohen, 2001 File divided into pieces Each recipient
More informationINTERNATIONAL JOURNAL OF PURE AND APPLIED RESEARCH IN ENGINEERING AND TECHNOLOGY
INTERNATIONAL JOURNAL OF PURE AND APPLIED RESEARCH IN ENGINEERING AND TECHNOLOGY A PATH FOR HORIZING YOUR INNOVATIVE WORK PEER-TO-PEER FILE SHARING WITH THE BITTORRENT PROTOCOL APURWA D. PALIWAL 1, PROF.
More informationCooperative End-to-end content distribution. Márk Jelasity
Cooperative End-to-end content distribution Márk Jelasity Content distribution So far we looked at search Content distribution is about allowing clients (peers) to actually get a file or other data after
More informationOPTIMIZATION QUESTIONS IN BITTORRENT COMMUNITIES
OPTIMIZATION QUESTIONS IN BITTORRENT COMMUNITIES Elvira Antal Department of Natural Sciences and Engineering, Faculty of Mechanical Engineering and Automation, E-mail: antal.elvira@gamf.kefo.hu Abstract
More informationA Fuzzy Approach For Piece Selection In Bit Torrent Like Peer To Peer Network. * M. Padmavathi, ** R. M. Suresh
A Fuzzy Approach For Piece Selection In Bit Torrent Like Peer To Peer Network * M. Padmavathi, ** R. M. Suresh * Assistant Professor, Department of Information Technology, RMD Engineering College, PO box
More informationBitTorrent Fairness Analysis
BitTorrent Fairness Analysis Team Asians Zhenkuang He Gopinath Vasalamarri Topic Summary Aim to test how the fairness affect the file transfer speed in a P2P environment (here using the BitTorrent Protocol)
More informationGame Theory. Presented by Hakim Weatherspoon
Game Theory Presented by Hakim Weatherspoon Game Theory Main Question: Can we cheat (and get away with it)? BitTorrent P2P file distribution tool designed with incentives for contribution Users contribute
More informationCS 3516: Advanced Computer Networks
Welcome to CS 3516: Advanced Computer Networks Prof. Yanhua Li Time: 9:00am 9:50am M, T, R, and F Location: Fuller 320 Fall 2017 A-term 1 Some slides are originally from the course materials of the textbook
More informationPeer-to-Peer Applications Reading: 9.4
Peer-to-Peer Applications Reading: 9.4 Acknowledgments: Lecture slides are from Computer networks course thought by Jennifer Rexford at Princeton University. When slides are obtained from other sources,
More informationCS5412: TORRENTS AND TIT-FOR-TAT
1 CS5412: TORRENTS AND TIT-FOR-TAT Lecture VII Ken Birman BitTorrent 2 Widely used download technology Implementations specialized for setting Some focus on P2P downloads, e.g. patches Others focus on
More informationDissecting BitTorrent: Five Months in a Torrent s Lifetime
Dissecting BitTorrent: Five Months in a Torrent s Lifetime M. Izal, G. Urvoy-Keller, E.W. Biersack, P.A. Felber, A. Al Hamra, and L. Garcés-Erice Institut Eurecom, 2229, route des Crêtes, 694 Sophia-Antipolis,
More informationUnderstanding BitTorrent: An Experimental Perspective
Understanding BitTorrent: An Experimental Perspective Arnaud Legout, Guillaume Urvoy-Keller, Pietro Michiardi To cite this version: Arnaud Legout, Guillaume Urvoy-Keller, Pietro Michiardi. Understanding
More informationP2P content distribution
P2P content distribution T-110.7100 Applications and Services in Internet, Fall 2010 Jukka K. Nurminen 1 V1-Filename.ppt / yyyy-mm-dd / Initials Steps of content sharing Share content Find content Transfer
More informationA Further Look at the Distance-Availability Weighted Piece Selection Method
473 A Further Look at the Distance-Availability Weighted Piece Selection Method A BitTorrent Piece Selection Method for On-Demand Media Streaming Petter Sandvik and Mats Neovius Department of Information
More informationAnalyzing and Improving a BitTorrent Network s Performance Mechanisms
Analyzing and Improving a BitTorrent Network s Performance Mechanisms Ashwin R. Bharambe Cormac Herley Venkata N. Padmanabhan Carnegie Mellon University Microsoft Research Microsoft Research Abstract In
More informationCache Capacity Allocation to Overlay Swarms
Cache Capacity Allocation to Overlay Swarms Ioanna Papafili 1, George D. Stamoulis 1, Frank Lehrieder 2, Benjamin Kleine 2, Simon Oechsner 2, 1 Athens University of Economics and Business, Athens, Greece
More informationBitTorrent and CoolStreaming
BitTorrent and CoolStreaming Jukka K. Nurminen Data Communications Software (DCS) Lab, Department of Computer Science and Engineering, Aalto University Jukka K. Nurminen Aalto University P2P Networks BitTorrent
More informationPeer to Peer Systems and Probabilistic Protocols
Distributed Systems 600.437 Peer to Peer Systems & Probabilistic Protocols Department of Computer Science The Johns Hopkins University 1 Peer to Peer Systems and Probabilistic Protocols Lecture 11 Good
More informationDepartment of Computer Science Institute for System Architecture, Chair for Computer Networks. File Sharing
Department of Computer Science Institute for System Architecture, Chair for Computer Networks File Sharing What is file sharing? File sharing is the practice of making files available for other users to
More informationApplication-Layer Protocols Peer-to-Peer Systems, Media Streaming & Content Delivery Networks
COMP 431 Internet Services & Protocols Application-Layer Protocols Peer-to-Peer Systems, Media Streaming & Content Delivery Networks Jasleen Kaur February 14, 2019 Application-Layer Protocols Outline Example
More informationP2P Applications. Reti di Elaboratori Corso di Laurea in Informatica Università degli Studi di Roma La Sapienza
P2P Applications Reti di Elaboratori Corso di Laurea in Informatica Università degli Studi di Roma La Sapienza Versione originale delle slides fornita da Dora Spenza e Marco Barbera P2P Paradigm Late 80
More informationChapter 2: Application layer
Chapter 2: Application layer 2. Principles of network applications app architectures app requirements 2.2 Web and HTTP 2.4 Electronic Mail SMTP, POP3, IMAP 2.5 DNS 2.6 P2P applications 2.7 Socket programming
More informationUnderstanding BitTorrent: An Experimental Perspective
Understanding BitTorrent: An Experimental Perspective Arnaud Legout, Guillaume Urvoy-Keller, Pietro Michiardi To cite this version: Arnaud Legout, Guillaume Urvoy-Keller, Pietro Michiardi. Understanding
More informationOn Randomness in ISP-friendly P2P Applications
2012 26th IEEE International Conference on Advanced Information Networking and Applications On Randomness in ISP-friendly P2P Applications S M Saif Shams Simula Research Lab University of Oslo Oslo, Norway
More informationApplication Layer: P2P File Distribution
Application Layer: P2P File Distribution EECS 3214 Slides courtesy of J.F Kurose and K.W. Ross, All Rights Reserved 29-Jan-18 1-1 Chapter 2: outline 2.1 principles of network applications 2.2 Web and HTTP
More informationDistributed Data Management. Christoph Lofi Institut für Informationssysteme Technische Universität Braunschweig
Distributed Data Management Christoph Lofi Institut für Informationssysteme Technische Universität Braunschweig http://www.ifis.cs.tu-bs.de 8.0 Content Provisioning 8.0 Content Distribution 8.1 Swarming
More informationReti P2P per file distribution: BitTorrent
Reti P2P per file distribution: BitTorrent Corso di Applicazioni Telematiche A.A. 2008-09 Lezione n.22 Prof. Roberto Canonico Università degli Studi di Napoli Federico II Facoltà di Ingegneria P2p file-sharing
More informationImproving Peer-to-Peer File Distribution: Winner Doesn t Have to Take All
Improving Peer-to-Peer File Distribution: Winner Doesn t Have to Take All ABSTRACT Ben Leong, Youming Wang, Su Wen, Cristina Carbunaru, Yong Meng Teo National University of Singapore 13 Computing Drive
More informationImproving the Download Time of BitTorrent-like Systems
Improving the Download Time of BitTorrent-like Systems Chi-Jen Wu, Cheng-Ying Li, and Jan-Ming Ho Institute of Information Science Academia Sinica, Taiwan {cjwu, cyli, hoho}@iis.sinica.edu.tw Abstract
More informationCSE 486/586 Distributed Systems Peer-to-Peer Architectures
CSE 486/586 Distributed Systems eer-to-eer Architectures Steve Ko Computer Sciences and Engineering University at Buffalo CSE 486/586 Last Time Gossiping Multicast Failure detection Today s Question How
More informationCMSC 332 Computer Networks P2P and Sockets
CMSC 332 Computer Networks P2P and Sockets Professor Szajda Announcements Programming Assignment 1 is due Thursday Where are we? What sorts of problems are we having? 2 Recap SMTP is the language that
More informationLast Lecture SMTP. SUNY at Buffalo; CSE 489/589 Modern Networking Concepts; Fall 2010; Instructor: Hung Q. Ngo 1
Last Lecture SMTP SUNY at Buffalo; CSE 489/589 Modern Networking Concepts; Fall 2010; Instructor: Hung Q. Ngo 1 This Lecture Peer-to-Peer (P2P) Applications SUNY at Buffalo; CSE 489/589 Modern Networking
More informationBitTorrent for Really Selfish Peers T4T Trading over Cycles
Semester Thesis BitTorrent for Really Selfish Peers T4T Trading over Cycles Roger Odermatt oroger@student.ethz.ch Professor: Dr. Roger Wattenhofer Advisor: Raphael Eidenbenz Department of Information Technology
More informationUniversity of Würzburg Institute of Computer Science Research Report Series
University of Würzburg Institute of Computer Science Research Report Series Pushing the Performance of Biased Neighbor Selection through Biased Unchoking Simon Oechsner 1, Frank Lehrieder 1, Tobias Hoßfeld
More informationUTAPS: An Underlying Topology-aware Peer Selection Algorithm in BitTorrent *
22nd International Conference on Advanced Information Networking and Applications UTAPS: An Underlying Topology-aware Peer Selection Algorithm in BitTorrent * Wei LI 1, Shanzhi CHEN 1, 2, Tao YU 1 ( 1
More informationP2P. 1 Introduction. 2 Napster. Alex S. 2.1 Client/Server. 2.2 Problems
P2P Alex S. 1 Introduction The systems we will examine are known as Peer-To-Peer, or P2P systems, meaning that in the network, the primary mode of communication is between equally capable peers. Basically
More informationarxiv:cs.ni/ v1 21 Nov 2006
Clustering and Sharing Incentives in BitTorrent Systems Arnaud Legout Nikitas Liogkas Eddie Kohler Lixia Zhang I.N.R.I.A. University of California, Los Angeles Sophia Antipolis, France Los Angeles, CA,
More informationConnectivity properties of real BitTorrent swarms
KSII TRANSACTIONS ON INTERNET AND INFORMATION SYSTEMS VOL. 7, NO. 9, Sep. 2013 2246 Copyright c 2013 KSII Connectivity properties of real BitTorrent swarms Ruben Cuevas 1, Michal Kryczka 12, Angel Cuevas
More information416 Distributed Systems. Mar 3, Peer-to-Peer Part 2
416 Distributed Systems Mar 3, Peer-to-Peer Part 2 Scaling Problem Millions of clients server and network meltdown 2 P2P System Leverage the resources of client machines (peers) Traditional: Computation,
More informationOn maximum throughput in BitTorrent
Gradus Vol 3, No 2 (2016) 67-72 ISSN 2064-8014 On maximum throughput in BitTorrent Elvira Dobjánné Antal 1, and Tamás Vinkó 2 1 Department of Natural Sciences and Engineering, Faculty of Mechanical Engineering
More informationCS 3516: Computer Networks
Welcome to CS 3516: Computer Networks Prof. Yanhua Li Time: 9:00am 9:50am M, T, R, and F Location: AK219 Fall 2018 A-term 1 Some slides are originally from the course materials of the textbook Computer
More informationBitTorrent. Internet Technologies and Applications
BitTorrent Internet Technologies and Applications Aims and Contents Aims Show benefits of P2P for file downloads Explain concepts of BitTorrent as a P2P file sharing application Show how BitTorrent protocol
More informationDelft University of Technology Parallel and Distributed Systems Report Series. Bandwidth Allocation in BitTorrent-like VoD Systems under Flashcrowds
Delft University of Technology Parallel and Distributed Systems Report Series Bandwidth Allocation in BitTorrent-like VoD Systems under Flashcrowds Lucia D Acunto, Tamás Vinkó, Henk Sips ldacunto@tudelftnl
More informationUnderstanding and Deconstructing BitTorrent Performance
Understanding and Deconstructing BitTorrent Performance Ashwin R. Bharambe Carnegie Mellon University Cormac Herley Venkata N. Padmanabhan Microsoft Research January 25 Technical Report MSR-TR-25-3 Microsoft
More informationThe Importance of History in a Media Delivery System
The Importance of History in a Media Delivery System Richard J. Dunn, Steven D. Gribble, Henry M. Levy, John Zahorjan University of Washington E-mail: {rdunn,gribble,levy,zahorjan}@cs.washington.edu Abstract
More informationLecture 8: Application Layer P2P Applications and DHTs
Lecture 8: Application Layer P2P Applications and DHTs COMP 332, Spring 2018 Victoria Manfredi Acknowledgements: materials adapted from Computer Networking: A Top Down Approach 7 th edition: 1996-2016,
More informationModeling, Analysis and Improvement for BitTorrent-Like File Sharing Networks
Modeling, Analysis and Improvement for BitTorrent-Like File Sharing etworks Ye Tian, Di Wu, and Kam Wing g Department of Computer Science and Engineering The Chinese University of Hong Kong Shatin,.T.,
More informationENSC 427: COMMUNICATION NETWORKS
ENSC 427: COMMUNICATION NETWORKS Comparison of TCP with "utp" for BitTorrent transfers Spring 2009 FINAL PROJECT Adam Ciapponi Robert Hueber Robert Szolomicki aciappon@sfu.ca rhueber@sfu.ca rms6@sfu.ca
More informationWho Profits from Peer-to-Peer File-Sharing? Traffic Optimization Potential in BitTorrent Swarms
Who Profits from Peer-to-Peer File-Sharing? Traffic Optimization Potential in BitTorrent Swarms Valentin Burger, Frank Lehrieder, Tobias Hoßfeld, Jan Seedorf University of Würzburg, Institute of Computer
More informationAVALANCHE: A NETWORK CODING ANALYSIS
COMMUNICATIONS IN INFORMATION AND SYSTEMS c 2007 International Press Vol. 7, No. 4, pp. 353-358, 2007 003 AVALANCHE: A NETWORK CODING ANALYSIS RAYMOND W. YEUNG Abstract. In this paper, we study the application
More informationDesign Space Analysis for Modeling Incentives in Distributed Systems
Design Space Analysis for Modeling Incentives in Distributed Systems by Rameez Rahman, Tamas Vinko, David Hales, Johan Pouwelse, and Henk Sips Delft University of Technology 1 Incentives in Distributed
More informationBitRiver: Final Report
BitRiver: Final Report Jonathan Stiansen, Sampoorna Biswas, Wali Usmani April 24, 2015 Abstract Peer-to-peer multimedia sharing has become widely popular due to its robust, fault-tolerant nature. We propose
More informationThe Design Trade-offs of BitTorrent-like File Sharing Protocols
The Design Trade-offs of BitTorrent-like File Sharing Protocols Bin Fan John C.S. Lui Dah-Ming Chiu Abstract The BitTorrent BT file sharing protocol is very popular due to its scalability property and
More informationDistributed Data Management
Distributed Data Management Wolf-Tilo Balke Christoph Lofi Institut für Informationssysteme Technische Universität Braunschweig http://www.ifis.cs.tu-bs.de 8.0 Content Provisioning 8.0 Content Distribution
More informationTraffic Patterns in Peer-to-Peer-Networking. Christian Schindelhauer. joint work with Amir Alsbih Thomas Janson
Traffic Patterns in Peer-to-Peer-Networking Christian Schindelhauer joint work with Amir Alsbih Thomas Janson to be presented at ITA Albert-Ludwig University Freiburg Department of Computer Science Computer
More informationUsing Home Routers as Proxies for Energy-Efficient BitTorrent Downloads to Mobile Phones
CONSUMER COMMUNICATIONS AND NETWORKING Using Home Routers as Proxies for Energy-Efficient BitTorrent Downloads to Mobile Phones Imre Kelényi and Ákos Ludányi, Budapest University of Technology and Economics
More informationCombining Coding and Block Schemes for P2P Transmission
Combining Coding and Block Schemes for P2P Transmission Ben Dodson bdodson@seas.upenn.edu Dr. Zach Ives ives@cis.upenn.edu Doug Petkanics dob@seas.upenn.edu Dr. Sanjeev Khanna sanjeev@cis.upenn.edu April
More informationBitTorrent from swarms to collectives
BitTorrent from swarms to collectives Some on-going research in the tribler team at TU Delft name: David Hales date: Sept. 21, 2009 event: ECCS 2009 Tribler team: 35 + collaborators tribler.org Tribler
More informationFurther extensions to the Wigan architecture are discussed in this thesis, including possible means of handling data updates.
COMPUTING SCIENCE A BitTorrent-based Peer-to-Peer Database Server J. Colquhoun TECHNICAL REPORT SERIES No. CS-TR-1135 January, 2009 TECHNICAL REPORT SERIES No. CS-TR-1135 January, 2009 A BitTorrent-based
More informationSmall Is Not Always Beautiful
Small Is Not Always Beautiful Paweł Marciniak Poznan University of Technology, Poland pawel.marciniak@gmail.com Nikitas Liogkas UCLA Los Angeles, CA nikitas@cs.ucla.edu Arnaud Legout I.N.R.I.A. Sophia
More informationARTICLE IN PRESS. Journal of Network and Computer Applications
Journal of Network and Computer Applications 32 (29) 666 683 Contents lists available at ScienceDirect Journal of Network and Computer Applications journal homepage: www.elsevier.com/locate/jnca Principles
More informationinternet technologies and standards
Institute of Telecommunications Warsaw University of Technology 25 internet technologies and standards Piotr Gajowniczek Andrzej Bąk Michał Jarociński Internet application layer peer-to-peer systems overview
More informationLecture 21 P2P. Napster. Centralized Index. Napster. Gnutella. Peer-to-Peer Model March 16, Overview:
PP Lecture 1 Peer-to-Peer Model March 16, 005 Overview: centralized database: Napster query flooding: Gnutella intelligent query flooding: KaZaA swarming: BitTorrent unstructured overlay routing: Freenet
More informationCooperation in Open Distributed Systems. Stefan Schmid
Cooperation in Open Distributed Systems Stefan Schmid T-Labs, Berlin, July 2, 2009 Distributed Systems 2008/9 Wireless: Many mobile phones today have WLAN (and even Skype) P2P: Olympic games 2008 live-broadcast
More informationT-Chain: A General Incentive Scheme for Cooperative Computing
: A General Incentive Scheme for Cooperative Computing Kyuyong Shin, Carlee Joe-Wong, Sangtae Ha, Yung Yi, Injong Rhee, Douglas Reeves Korea Military Academy, Princeton University, University of Colorado
More informationimproving the performance and robustness of P2P live streaming with Contracts
MICHAEL PIATEK AND ARVIND KRISHNAMURTHY improving the performance and robustness of P2P live streaming with Contracts Michael Piatek is a graduate student at the University of Washington. After spending
More informationA Method of Identifying the P2P File Sharing
IJCSNS International Journal of Computer Science and Network Security, VOL.10 No.11, November 2010 111 A Method of Identifying the P2P File Sharing Jian-Bo Chen Department of Information & Telecommunications
More informationUnderstanding BitTorrent Through Real Measurements
Understanding BitTorrent Through Real Measurements Wojciech Mazurczyk, Paweł Kopiczko Warsaw University of Technology, Institute of Telecommunications Warsaw, Poland, -665, Nowowiejska 15/19 Email: wmazurczyk@tele.pw.edu.pl,
More informationA Measurement Study of Attacks on BitTorrent Seeds
A Measurement Study of Attacks on BitTorrent Seeds Prithula Dhungel, Xiaojun Hei,DiWu, Keith W. Ross Polytechnic Institute of NYU, Brooklyn, NY 1121 Huazhong University of Science and Technology, P. R.
More informationPeer-to-Peer Media Streaming
SEMINAR: RECHNERNETZE UND VERTEILTE SYSTEME HS08 Peer-to-Peer Media Streaming Francisco Piña Garrido University Jaen (Spain) November 26th, 2008 Overview > Introduction Definition of Peer to Peer Streaming
More informationA Novel ALTO Scheme for BitTorrent-Like P2P File Sharing Systems
2013 Third International Conference on Intelligent System Design and Engineering Applications A Novel ALTO Scheme for BitTorrent-Like P2P File Sharing Systems Liu Guanxiu, Ye Suqi, Huang Xinli Department
More informationA MASSIVELY SCALABLE PERSISTENT CONTENT DISTRIBUTION SYSTEM
A MASSIVELY SCALABLE PERSISTENT CONTENT DISTRIBUTION SYSTEM ABSTRACT This paper proposes a novel form of peercasting system as an improved solution for IP-based mass media content delivery. Several approaches
More informationPeer-to-Peer Networks 12 Fast Download
Peer-to-Peer Networks 12 Fast Download Arne Vater Technical Faculty Computer Networks and Telematics University of Freiburg IP Multicast Motivation - Transmission of a data stream to many receivers Unicast
More informationIndian Journal of Science and Technology Vol. 5 No. 7 (July 2012) ISSN:
A novel approach for mitigating Distributed Denial of Service attacks drawn on bit-torrent protocol in computer networks S. S. Nagamuthu Krishnan 1* and V. Saravanan 2 1 Thiagarajar School of Management,
More informationINVESTIGATING AND IMPROVING BITTORRENT S PIECE AND NEIGHBOR SELECTION ALGORITHMS
INVESTIGATING AND IMPROVING BITTORRENT S PIECE AND NEIGHBOR SELECTION ALGORITHMS by Cameron Dale B.Sc., Simon Fraser University, 2000 a thesis submitted in partial fulfillment of the requirements for the
More informationChanging the Unchoking Policy for an Enhnaced BitTorrent
1/1 Changing the Unchoking Policy for an Enhnaced BitTorrent Vaggelis Atlidakis, Mema Roussopoulos and Alex Delis Department of Informatics and Telecommunications, University of Athens, 15748, Greece April
More information